TRANSITIONAL JOBS
RETURN ON INVESTMENT STUDY
Trudi Renwick
Senior Economist
August 2008
Transitional Jobs programs (TJ) are a practical workforce strategy, using time-limited, wagepaying jobs which combine real work, skill development, and supportive services to transition
participants rapidly and successfully into the labor market. The essential underpinning of these programs
is that every participant can be successful in the workplace. TJ programs reflect a flexible, evolving
design comprised of core fundamental principles combined with elements tailored to the target
population. TJ programs can vary as to length, the type of transitional employer (for profit, nonprofit, or
government), the role of formal skills training, funding streams, and the population served. TJ programs
can be delivered in an urban or rural context of a micro-enterprise, a publicly or privately funded job, or
work crews bidding on jobs.
In any of these settings, the following are key common elements:
Transitional Jobs programs offer a model of employment by which participants learn by doing –
they experience the customs and routines of work, acquire work-task skills, establish an
employment record, and generate employer references to enhance their competitiveness in private
sector employment.
Transitional Jobs programs assume that everyone can work. With confidence gained from success
in the workplace and appropriate case management to assisting with any problems along the way,
TJ programs provide the participant with learning opportunities, pay, and the support necessary to
transition to full-time, permanent employment.
This return on investment study is an attempt to quantify the public cost savings which can be
expected to result from a TJ initiative in New York State aimed at two specific target audiences: public
assistance recipients (those who are receiving public assistance and have been unemployed for more than
12 consecutive months) and formerly incarcerated individuals (those who have returned from jail or
prison within the previous 90 days). The analysis will compare the cost of providing transitional jobs
(including the cost of wage payments, training, supervision and other program administration) with the
public savings that result when participants increase their likelihood of employment and therefore are
more likely to stay out of jail or prison and get off public cash assistance rolls.
Transitional Jobs Initiative Parameters
This analysis considers the costs and cost savings related to a statewide TJ initiative consisting of
4,000 publicly-financed transitional job "slots" over one year. The pay for these transitional jobs is
set at $8.00 per hour for a 35-hour workweek.
On average, participants are expected to stay in a TJ program for six months. Therefore the statewide
initiative will enroll, over the course of 12 months, approximately 8,000 different individuals.
1
This statewide initiative will have two primary target populations: public assistance recipients (those
who are receiving public assistance and have been unemployed for more than 12 consecutive months)
and formerly incarcerated individuals (those who have returned from jail or prison with previous 90
days). The return on investment analysis assumes that approximately half of the 8,000 participants
each year will come from each group.
Transitional Jobs Program Investment Requirements - Direct, Net and Overall
Table 1 summarizes the "direct" costs of the initiative, including wages, supervision, training and
administrative costs. Costs for the supervision, training and the administration of the initiative are
estimated at $3,000 per participant over six months or $6,000 per slot per year. The $3,000 per
participant estimate is based on an analysis of the administrative costs of the New York City Parks and
Recreation Intra-city Park Opportunity Program (POP). The TJ initiative would spend the same $3,000
per participant but over a shorter six-month period, providing twice as much support to each participant
each month as the POP program.
Table 1: Direct Cost of Transitional Jobs Initiative
Wages @$8.00 per hour - 35 hours per week
Administration, Training and Supervision
Total Monthly Cost per Participant
Total Annual Cost for 8,000 participants - six
month participation per participant
$1,213
$500
$1,713
$82,224,000
Table 2 summarizes the "net" costs of the initiative. Since three quarters of the participants will
be recruited from existing public assistance recipients (all the public assistance recipients and half the
formerly incarcerated), earnings from the TJ program will replace cash assistance and these participants
will no longer be participating in other work-related activities financed by the local social services
district. This reduces the monthly cost per participant for public assistance recipients from the $1,713 per
month to $705 or $982 per month (depending on family size) and the total annual program cost from $82
million to $49 million. If less than half the formerly incarcerated participants are receiving cash
assistance at the time of enrollment, the net state/local cost of the program will be slightly higher.
Table 2: Net State/Local Cost of Transitional Jobs Initiative
Formerly Incarcerated
Public Assistance Receiving Cash
Not Receiving Cash
Recipients
Assistance at Time Assistance at Time of
of Enrollment
Enrollment
Number of Participants
Total Monthly Cost per Participant
Public Assistance Cost Savings (Cash
4,000
$1,713
2,000
$1,713
2,000
$1,713
Grant and Workfare Administration)
$1,008
$705
$731
$982
$0
$1,713
$16,920,000
$11,784,000
$49,260,000
$20,556,000
Net Monthly Cost per Participant
Annual Net Cost
Total Annual Net Cost
2
Table 3 provides a broader picture of the public costs associated with the TJ initiative, examining
not just public assistance outlays and savings but taking into account costs of tax credits, food stamps,
child care, public health insurance programs and differentiating between federal and state/local financing
responsibility for each item. For public assistance recipients, the outlays for child care, transportation and
Medicaid would be identical to the expenditures for these purposes made on behalf of public assistance
recipients participating in mandated work-related programs. For participants who had not been receiving
public assistance at the time of enrollment in the program, enrollment in the program will increase
expenditures in these areas.
Table 3: Total Public Cost for Transitional Jobs Program Participant
Parent with two children
(one pre-school and one school-age)
Total
Wages @$8.00 per hour - 35
hours per week
Food Stamps
Federal EITC
State EITC
City EITC
Child Care
Transportation
Medicaid
Transitional Jobs Admin
TOTAL MONTHLY COST
TOTAL SIX MONTH COST
$1,213
$304
$402
$121
$20
$1,174
$81
$1,062
$500
$4,877
$29,262
Federal
$0
$304
$402
$0
$0
$0
$0
$531
$0
$1,237
$7,422
State/Local
$1,213
$0
$0
$121
$20
$1,174
$81
$531
$500
$3,640
$21,840
Single individual with no dependents
Total
$1,213
$0
$0
$0
$0
$0
$81
$556
$500
$2,350
$14,100
Federal
$0
$0
$0
$0
$0
$0
$0
$278
$0
$278
$1,668
State/Local
$1,213
$0
$0
$0
$0
$0
$81
$278
$500
$2,072
$12,432
Public Cost of Alternative Outcomes
In order to estimate the return on investment, this analysis estimates the state/local public costs
associated with four different outcomes: (1) the continued receipt of public assistance; (2) incarceration;
(3) unsubsidized employment at the minimum wage of $7.25 per hour for 1,500 hours per year; and (4)
unsubsidized full-time, full-year employment at $10.00 per hour. 1
1
Appendix A includes a list of 29 job titles in which TJ participants are likely to be placed. The median starting
wage for these positions is $10.27 an hour. To the extent that participants’ wages will increase over time in these
jobs, the average wage for participants who complete the training and are placed in unsubsidized employment may
be even greater than $10 an hour. Higher average wages would result in even greater reductions in public costs.
3
Public Assistance
Public assistance costs, summarized in Table 4, are based on the public assistance standard of need for a
three-person family (unmarried parent with two children) and the public assistance standard of need for a
single person in New York City (NYC).2 The cash assistance estimate assumes no earned income for the
family and that the family lives in an apartment with heat included in the rent with a rent equal to the
NYC shelter allowance. The estimate assumes that the adult in the family participates in a work program3
and therefore is eligible for and receives transportation assistance4 and free childcare. 5 In addition, the
estimate assumes that the family is covered by Medicaid6 and receives food stamps. For a single
individual, the cash assistance estimate is based on a single person case and does not include any
provision for childcare.
Table 4: Monthly Public Cost for Public Assistance Recipient
Parent with two children (one pre-school
Single individual with no dependents
and one school-age)
Total
Cash Assistance
Food Stamps
Child Care
Transportation
Medicaid
Workfare Administration
MONTHLY TOTAL
$691
$388
$1,174
$81
$1,062
$317
$3,713
Federal
$0
$388
$0
$0
$531
$0
$919
State/Local
$691
$0
$1,174
$81
$531
$317
$2,794
Total
$414
$162
$0
$81
$556
$317
$1,530
Federal
$0
$162
$0
$0
$278
$0
$440
State/Local
$414
$0
$0
$81
$278
$317
$1,090
2
This analysis estimates public assistance costs using the New York City standard of need. The standard of need
consists of the basic allowance, the home energy allowance and the supplemental home energy allowance that are
the same across the state and the shelter allowance that varies by social services district. The shelter allowance for a
three-person family with children ranges from $259 in Franklin County to $447 in Suffolk County. New York
City's Shelter Allowance for a three-person family with children is $400. Since the majority of the public assistance
recipients are in New York City, it is appropriate to use the New York City standard of need in these calculations.
Public Assistance savings per participant will be somewhat higher in social service districts with shelter allowances
greater than New York City (Westchester, Nassau, Suffolk, Nassau, Putnam, Rockland, Dutchess and Orange) and
slightly lower in other upstate counties.
3
Workfare administration at $3,802 annually is set equal to average base rates for Back to Work Contracts.
4
The $81 cost of transportation assumes a 30-day unlimited MTA pass.
5
Child care costs are based on the Registered Family Day Care Market Rates published by the Office of Children
and Family Services: $160 per week for preschool for twelve months, $105 per week for school age for 10 months
and $141 per week for school age for the two months of summer.
6
Medicaid is based on average expenditure per participant for 2007 for NYC: $3,026 per kid; $6,682 per adult.
4
Incarceration
According to the New York State Department of Correctional Services, New York spends $32,426 per
inmate per year.7 This estimate includes only the operating costs of incarceration.8 For the purposes of
this analysis, all incarceration costs are considered state/local costs. If individuals are incarcerated in
federal prisons, the costs would be federal rather than state or local.
Table 5: Public Cost of Incarceration
Annual Cost
Administration
Support Services
Security
Health Services
Program Services
TOTAL
Monthly Cost
$375.00
$6,439.00
$17,859.00
$4,661.00
$3,092.00
$32,426.00
$31.25
$536.58
$1,488.25
$388.42
$257.67
$2,702.17
Source: Office of Public Information, New York State Department of Correctional Services
Employment
Given the structure of New York's system of work supports, the public cost of employment depends
primarily on three factors: (1) income (which is itself a function of the hourly wage and the hours
worked); (2) whether or not the employer provides health insurance; and (3) whether or not the family
utilizes public child care subsidies. If income is low, either because wages are very low and/or the adult
works less than full time, full year, the household may still be eligible for cash assistance as well as food
stamps, the earned income tax credit, subsidized child care and public health insurance (either through
Medicaid, Family Health Plus or Child Health Plus). Even many households with higher earnings ($10
per hour working full time, full year) are eligible for food stamps, subsidized child care and public health
insurance but not all eligible families apply for and receive these benefits. Child care "take up" rates are
particularly low; a recent report estimated that only 30 percent of New York families eligible for
subsidized child care actually receive these subsidies.9
In order to get a reasonable estimate of the public costs associated with employment at different wages
and hours of work for a three-person family, the following two tables estimate the public cost of two
scenarios: full time, full year work at $10 per hour and 1,500 hours of work per year at $7.25 per hour
(the minimum wage effective July 24, 2009). In order to account for the fact that not all these workers
will receive public health insurance and child care subsidies, the assumption is made that the households
are uniformly distributed among four possibilities: (1) public health insurance and child care subsidy; (2)
public health insurance but no child care subsidy; (3) employer-provided health insurance and a child care
7
Erik Kriss, New York State Department of Correctional Services, Office of Public Information.
This is very conservative estimate of the cost of incarceration in New York. According to the U.S. Department of
Justice Bureau of Justice Statistics, in 2001 New York spent $36,835 per year per inmate to operate its state prisons.
Using the same price index for state and local government services used by the Department of Justice in it analysis
to update that 2001 estimate, the 2008 per inmate cost would be $50,383.
9
Randy Albeda and Heather Boushey, Bridging the Gaps: A Picture of How Work Supports Work in Ten States,
Center for Economic and Policy Research, October 2007.
8
5
subsidy; and (4) employer-provided health insurance and no child care subsidy. This is the equivalent to
an assumption that half the families use public health insurance programs and that half have subsidized
child care. This assumption is consistent with the fact that about half of the private sector workers in New
York have access to employer-provided health care and that about 30 percent of those eligible for child
care subsidies actually receive them.
For single individuals, the estimates are simpler. Working full-time, full year at $10 per hour, a single
worker without dependents is not eligible for any work supports. Even working 1,500 hours per year at
$7.25 per hour, a single individual is eligible for only $114 per month in food stamps that are federally
funded. Neither single worker would be eligible for Family Health Plus or the Federal EITC. If the
single worker is a noncustodial parent, current on his/her child support payments, he/she might be eligible
for New York State's EITC for noncustodial parents under either earnings scenario.
Table 6: Monthly Public Cost
for $7.25 per Hour, 1,500 Hour Per Year Worker
Parent with two children (one pre-school and one school-age)
Public Health Public Health Employer Employer Weighted Federal State/Local
Provided
Average
Insurance & Insurance but Provided
Child Care
no Child Health Ins. & Health Ins.
Subsidy Care Subsidy Child Care No Child
Subsidy
Care
Subsidy
Cash Assistance
$236
$236
$236
$236
$236
$0
$236
Food Stamps
$307
$307
$307
$307
$307 $307
$0
Federal EITC
$363
$363
$363
$363
$363 $363
$0
State EITC
$109
$109
$109
$109
$109
$0
$109
City EITC
$18
$18
$18
$18
$18
$0
$18
Child Care
$1,174
$0
$1,174
$0
$587
$0
$587
$756
$756
$0
$0
$378 $189
$189
$2,963
$1,789
$2,207
$1,033
$1,998 $859
$1,139
Family Health Plus/
Child Health Plus
TOTAL PUBLIC COST
6
Table 7: Monthly Public Cost
for $10 per Hour, Full Time, Full Year Worker
Parent with two children (one pre-school and one school-age)
Public
Public
Employer Employer Weighted Federal State/Local
Health
Health
Provided
Provided
Average
Insurance & Insurance but Health Ins. Health Ins.
Child Care no Child
& Child
No Child
Subsidy Care Subsidy
Care
Care
Subsidy
Subsidy
Food Stamps
$232
$232
$232
$232
$232
$232
$0
Federal EITC
$359
$359
$359
$359
$359
$359
$0
State EITC
$108
$108
$108
$108
$108
$0
$108
City EITC
$18
$18
$18
$18
$18
$0
$18
$1,024
$0
$1,024
$0
$512
$0
$512
$756
$756
$0
$0
$378
$189
$189
$2,497
$1,473
$1,741
$717
$1,607
$780
$827
Child Care
Family Health
Plus/Child Health
Plus
TOTAL
MONTHLY
PUBLIC COST
Assessment of the Likelihood of Each Outcome
The next step in this analysis involves assigning a likelihood to each of the outcomes for which cost
estimates have been made for the group participating in the TJ initiative and for a theoretical "control
group" which does not benefit from the TJ initiative. While these assessments are certainly subjective,
they are based on the data available assessing the outcomes of existing TJ programs for each population
group. The analysis concludes with an analysis of the impact of changing each of these assumptions on
the return on investment for each part of the initiative.
Our basic assumption is that participation in a TJ program doubles the employment rate for participants
and reduces the incarceration rate by one half. In addition, we assume that for those employed,
individuals who have participated in a TJ program are more likely to have a $10 per hour job than
individuals who have not participated in the program.
Our analysis does not assume that all TJ program participants will be "successful" – e.g. obtaining and
retaining full time, full year jobs paying $10 an hour and providing family health insurance benefits. Nor
does our analysis assume that individuals not participating in the program will never be "successful." We
do assume that participation in the program increases the likelihood of "successful" outcomes and
decreases the likelihood of continued receipt of public assistance and/or re-incarceration.
Assumptions for Public Assistance Recipients
For public assistance recipients, the assumption is that 60 percent of the participants in the six-month
program (including those who complete the program and those who enroll but do not complete the
program) will be employed in unsubsidized employment, half of these full time in $10 per hour jobs and
7
10
half in $7.25 per hour jobs for 1,500 hours per year. The average wage for the participants would be
$8.65 per hour, approximately 20 percent higher than the current state minimum wage. The average
number of hours worked for the participants would be 1,750 (average of 1,500 and 2,000). Therefore, the
11
average income gain for participants would be approximately 42 percent.
In both these cases, the
analysis assumes that half the participants with post-program employment have access to employerprovided health insurance and half receive a child care subsidy. We assume that the remaining 40 percent
of the participants will continue to receive public assistance.
Obviously, any particular participant may have some months in each of these categories and few would
be expected to remain in a single category for the entire 30 months covered by the analysis. Another way
to look at these assumptions is to consider them as a distribution not of the participants as much as a
distribution of the aggregate post-transitional jobs program months (30 times 4,000 participants = 120,000
months). In the aggregate, we expect that 36,000 months of employment at $10 per hour; 36,000 months
of employment at $7.25 an hour; 48,000 months receiving public assistance.12
In the primary analysis of this estimate, we assume that for the "control group" – an equal number of
public assistance recipients not participating in the transitional jobs program – only half as many are
employed (30 percent) and that only one third of those employed (10 percent) would have $10 per hour
jobs and two thirds would have the $7.25 per hour jobs.13 We assume 70 percent of the control group
would continue to receive public assistance. Again, it might be more useful to think of this in terms of
the distribution of the aggregate 120,000 months rather than assuming that participants would remain in a
single category for the entire three years.
These assumptions are summarized in Table 8 below:
Table 8: Assumptions Regarding Transitional Jobs Outcomes
for Public Assistance Recipients
Percent of
Participants
Percent of
Control Group
Employed at
$10 per Hour
Employed at
$7.25 per Hour
Public Assistance
Total
30.0%
30.0%
40.0%
100%
10.0%
20.0%
70.0%
100%
10
This is consistent with the outcomes found in studies of other TJ initiatives working with public assistance
recipients. Washington State's Community Jobs Program places 50 to 75 percent of participants into unsubsidized
jobs within 6 to 9 months of enrollment. (Burchfield and Yatsko, 2002). A similar study of the Georgia
GoodWorks! Program found job placement rates as high as 70 percent in the most successful program sites. (Derr,
2002).
11
This is consistent with the results of Washington State’s Community Jobs Program. In that program, participant
wages increased by an average of 60 percent during their first two years in the workforce and 148 percent relative to
their pre-participation income. (Burchfield and Yatsko, 2002). This study defined as "participant" any individual
who spent at least one day working in the Community Jobs program.
12
For example, the 36,000 months of employment at $10 an hour could be 1,200 participants working full time at
$10 an hour for the entire 30 months or (more likely) 2,000 participants employed for an average of 18 months each.
13
These assumptions are consistent with the summary statistics reported by Coppock (2007, p. 35) regarding the low
employment rates of current and former TANF recipients. "One study finds that unemployment rates among former
Illinois TANF recipients range from 48 to 62 percent." Since our control group is made up of TANF recipients, an
assumption of 30 percent employment is reasonable.
8
Assumptions for Formerly Incarcerated Individuals
Based on the outcomes documented for a TJ program working exclusively with formerly
incarcerated individuals, a similar set of assumptions was established for the second group of
participants. We assumed that unsubsidized employment rates would double,14 and reincarceration rates would be cut in half.15 The estimate of the 20 percent re-incarceration rate for
the control group is based on a study done by the New York State Department of Correctional
Services which found that, of the cohort of individuals released from New York State prisons in
2002, about 40 percent returned to prison after a median period of 18 months.16 (We therefore
assume that after 36 months, 40% of the control group would have spent half of the 36 months
back in prison.) These assumptions are summarized in Table 9.
Table 9: Assumptions Regarding Transitional Jobs Outcomes
for Formerly Incarcerated Individuals
Employed at
$10 per Hour
Employed at
$7.25 per Hour
Public
Assistance
Reincarcerated
Total
Percent of Participants
50%
30%
10%
10%
100%
Percent of Control Group
15%
25%
40%
20%
100%
Return on Investment
The cost estimates for each of the scenarios and the assumptions regarding the likelihood of each outcome
are combined in a final step to estimate a return on investment for the statewide TJ initiative. Table 10
and Table 11 provide a summary of the return on investment for the TJ initiative expenditures for each
group of participants. For public assistance recipients, an initial $17 million investment in the TJ
initiative results in state/local cost savings of $60 million dollars. For the formerly incarcerated, an initial
$32 million investment results in state/local cost savings of $96 million.
14
A 2007 MDRC report on the Center for Employment Opportunities transitional jobs program in New York City
found that program participants were twice as likely compared to the control group to be employed in all four
quarters of the first year of the study (22 percent vs. 11 percent). The MDRC study was not able to distinguish
between subsidized and unsubsidized employment. (http://www.mdrc.org/publications/468/full.pdf)
15
The MDRC study found that among participants recruited for the program (or the control group) within three
months of release from prison, incarceration rates were cut in half at the end of the first year.
(http://www.mdrc.org/publications/468/full.pdf)
16
"2002 Releases: Three Year Post Release Follow-up", New York State Department of Correctional Services,
2007.
9
Table 10: Public Assistance Recipients
Year 1: Initial Investment
Number of Participants
4,000
Annual Cost Net of State/Local Public Assistance Savings
$16,920,000
Employed at
$10 per Hour
Employed at
Public
$7.25 per Hour Assistance
Total
First Cohort of Participants Participating in
Program During Months 1 - 6
600
600
800
2,000
Control Group During Months 1 - 6
200
400
1,400
2,000
2,977,200
4,100,400
13,411,200
$20,488,800
992,400
2,733,600
23,469,600
$27,195,600
State/local Costs - Months 7 - 12 for First
Cohort of Participants
State/local Costs - Months 7 -12 for Control
Group
TOTAL STATE/LOCAL COST SAVINGS
generated YEAR 1
($6,706,800)
Year 2
Participants from First and Second Cohorts
Participating in Program in Year 1
Control Group in Year 1
State/local Costs – Year 2 for First and Second
Cohort of Participants
State/local Costs – Year 2 for Control Group
1,200
400
1,200
800
1,600
2,800
4,000
4,000
11,908,800
16,401,600
53,644,800
$81,955,200
3,969,600
10,934,400
93,878,400
$108,782,400
TOTAL STATE/LOCAL COST SAVINGS generated YEAR 2
($26,827,200)
Year 3
Participants from First and Second Cohorts
Participating in Program in Year 1
Control Group in Year 1
'State/local Costs – Year 3 for First and Second
Cohort of Participants
'State/local Costs – Year 3 for Control Group
1,200
400
1,200
800
1,600
2,800
4,000
4,000
11,908,800
16,401,600
53,644,800
$81,955,200
3,969,600
10,934,400
93,878,400
$108,782,400
TOTAL STATE/LOCAL COST SAVINGS generated YEAR 3
($26,827,200)
THREE YEAR CUMULATIVE COST SAVINGS
($60,361,200)
THREE YEAR NET COST ESTIMATE - INITIAL INVESTMENT MINUS COST SAVINGS
($43,441,200)
10
Table 11: Formerly Incarcerated Individuals
Year 1: Initial Investment
Number of Participants
4,000
Annual Cost Net of State/Local Public Assistance Savings
$32,340,000
Employed at Employed at Public
$10 per Hour $7.25 per Hour Assistance Incarcerated Total
First Cohort of Participants Participating in
Program During Months 1 - 6
Control Group During Months 1 - 6
State/local Costs - Months 7 - 12 for First
Cohort of Participants
State/local Costs - Months 7 -12 for
Control Group
1,000
600
200
200
2,000
300
500
800
400
2,000
$0
$0 $2,486,400
$3,242,596
$5,728,996
-
-
6,485,192
$16,430,792
9,945,600
TOTAL STATE/LOCAL COST SAVINGS generated YEAR 1
($10,701,796)
YEAR 2:
Participants from First and Second Cohorts
Participating in Program in Year 1
Control Group in Year 1
State/local Costs – Months 13 to 24 for
First and Second Cohort of Participants
2,000
1,200
400
400
4,000
600
1,000
1,600
800
4,000
-
$9,945,600 $12,970,384
$22,915,984
-
39,782,400
25,940,768
$65,723,168
($42,807,184)
1,200
400
400
4,000
1,000
1,600
800
4,000
-
$9,945,600 $12,970,384
$22,915,984
-
39,782,400
-
State/local Costs – Months 13 to 24 for
Control Group
TOTAL STATE/LOCAL COST SAVINGS generated YEAR 2
YEAR 3:
Participants from First and Second Cohorts
Participating in Program in Year 1
2,000
Control Group in Year 1
600
'State/local Costs – Months 25 to 36 for
First and Second Cohort of Participants
'State/local Costs – Months 25 to 36 for
Control Group
TOTAL STATE/LOCAL COST SAVINGS generated YEAR 3
25,940,768
$65,723,168
($42,807,184)
THREE YEAR CUMULATIVE COST SAVINGS
($96,316,164)
THREE YEAR NET COST ESTIMATE - INITIAL INVESTMENT MINUS COST SAVINGS
($63,976,164)
11
Impact of Changing Assumptions
The preceding analysis was based on a basic assumption that employment rates doubled for both
groups of participants (relative to the control group) and that recidivism rates were cut in half for
formerly incarcerated participants. Table 12 provides estimates of the cost savings with
alternative assumptions regarding employment and recidivism for each group including the
changes in employment and recidivism required to "break even", e.g. generate three-year cost
savings equal to the initial net program outlay.
Table 12: Three Year Net Cost Estimate Initial Investment Minus Cost Savings
Public Assistance Recipients
Change in Employment
100%
(from 30% to 60%)
50%
(from 30% to 45%)
26%
(from 30% to 38%)
25%
(from 30% to 37.5%)
10%
(from 30% to 33%)
($43,441,200)
($14,103,000)
$0
$566,100
$9,367,560
Formerly Incarcerated Individuals
Change in Recidivism
Change in Employment
100%
(from 40% to 80%)
50%
(from 40% to 60%)
35%
(from 40% to 54%)
32%
(from 40% to 53%)
29%
(from 40% to 52%)
25%
(from 40% to 50%)
10%
(from 40% to 44%)
50%
(from 40% to 20%)
25%
(from 40% to 30%)
10%
(from 40% to 36%)
($63,976,164)
($60,573,282)
($58,531,553)
($19,220,964)
($15,818,082)
($13,776,353)
($5,444,611)
($2,041,729)
$0
($3,402,882)
$0
$2,041,729
$0
$3,402,882
$5,444,611
$3,156,636
$6,559,518
$8,601,247
$16,583,196
$19,986,078
$22,027,807
Conclusion
This analysis has shown that a statewide TJ initiative in New York could clearly pay for itself in
just three years in primary public cost savings if it is able to increase employment rates by more
than 26% for public assistance recipients and by more than 35% for formerly incarcerated
participants. (If recidivism is reduced by more than 10%, then the minimum required percentage
increase in employment for formerly incarcerated participants is even smaller). And the actual
public cost savings as a result of this initiative would likely be even greater because many
12
additional public cost savings have not been incorporated into this analysis. Take, for instance,
the child support system. The formerly incarcerated participants would be expected to increase
their child support payments to their families, which would result in lower public assistance
payments to these families. Another example is in the area of justice system costs. This analysis
includes only the cost of incarceration, but there would also be significant public cost savings for
the entire judicial system with a reduction in recidivism. Furthermore, this is a conservative
estimate of the return on investment because we have not included in this analysis any of the
multiplier effects on the economy of the increased consumption capacity of the program
participants.
In an era of significant fiscal constraints at all levels of government, new initiatives should be
carefully evaluated for their long-term impact on the economy. This analysis has shown that an
investment of several million dollars in a statewide Transitional Jobs initiative would be an
excellent use of these scarce fiscal resources.
13
WORKS CITED
Burchfield, Erin and Sarah Yatsko (2002) From Welfare to Paycheck: Research on Barriers to
Work and Employment Outcomes of Washington State's Community Jobs Program (Report).
Seattle, WA: Economic Opportunity Institute.
http://www.eoionline.org/worker_advancement/reports/FromWelfareToPaycheckReportOct02.pdf
Bloom,Dan with Cindy Redcross, Janine Zweig (Urban Institute) and Gilda Azurdia. (November
2007) Transitional Jobs for Ex-Prisoners: Early Impacts from a Random Assignment Evaluation
of the Center for Employment Opportunities (CEO) Prisoner Reentry Program. (An MDRC
Working Paper). Manpower Development and Research Corporation.
Boushey, Heather and Randy Albeda (2007). Bridging the Gaps: A Picture of How Work
Supports Work in Ten States, Washington, DC: Center for Economic and Policy Research.
Coppock, Abigail (2007). "Transitional Jobs: Overcoming Barriers to Employment," Advocates
Forum, University of Chicago School of Social Service Administration.
Derr, Michelle K., LaDonna A. Pavetti, and Angelina Kewal Ramani (2002). Georgia Good
Works! Transitional Work and Intensive Support for TANF Recipients Nearing the
TimeLimit(Report). Washington, DC: Mathematica Policy Research. http://www.mathematicampr.com/publications/PDFs/goodworks.pdf
Florida Department of Corrections (June 1999). Return on Investment for Correctional
Education in Florida. Bureau of Research and Data Analysis, Florida Department of
Corrections.
National Transitional Jobs Network (2007). Evaluation of Transitional Jobs Programs.
National Transitional Jobs Network (2007). Transitional Jobs Programs for Persons with
Criminal Records.
New York State Department of Corrections (2007?). 2002 Releases: Three Year Post Release
Follow-up, 2007 (? - no date on publication).
http://www.docs.state.ny.us/Research/Reports/2002_Releases_3YR_OUT.pdf
Stephan, James J. (2004, June). State Prison Expenditures, 2001. Bureau of Justice Statistics
Special Report, Office of Justice Programs, U.S. Department of Justice.
Waller, Margy (2002,May). Transitional Jobs: A Next Step in Welfare to Work Policy, Center
on Urban & Metropolitan Policy, The Brookings Institute.
14
Appendix A
Unsubsidized Positions
Suggested in TJ Proposal
Childcare Worker
Teacher Assistant
Social and Human Service
Assistant
Recreation Worker
Customer Service Representative
Home Health Aide
Medical Assistant
Dental Assistant
Nursing Aide/Orderly/Attendant
Office Clerk
Secretary
Receptionist
Bookkeeping, Auditing,
Accounting Clerk
Word Processor/Typist
Bus Driver, School
Bus Driver, Transit and Intercity
Truck Driver, Light or Delivery
Truck Driver, Heavy or Tractor
Trailer
Construction Trade Worker
NYS DOL
Entry-Level Annual Wage
Entry-Level Hourly Wage
(=Annual/2,080 hours)
$16,670
$16,910
$8.01/hour
$8.13/hour
$19,880
$16,670
$21,360
$16,970
$23,190
$25,650
$21,770
$18,120
$22,510
$20,010
$9.56/hour
$8.01/hour
$10.27/hour
$8.16/hour
$11.15/hour
$12.33/hour
$10.47/hour
$8.71/hour
$10.82/hour
$9.62/hour
$23,910
$24,440
$23,140
$31,130
$18,030
$11.50/hour
$11.75/hour
$11.13/hour
$14.97/hour
$8.67/hour
$27,160
$13.06/hour
$24,230 - $37,260
$11.65/hour - $17.91/hour
$17,510 - $22,770
$8.42/hour - $10.95/hour
$17,790
$26,200
$31,550
$19,580
Not available
$8.55/hour
$12.60/hour
$15.17/hour
$9.41/hour
Not available
$33,560
$16.13/hour
$21,360
$10.27/hour
$18,680
$8.98/hour
$17,180
$8.26/hour
Construction Trades Helper
Landscaping and
Groundskeeping Worker
Highway Maintenance Worker
Tree Trimmer and Pruner
Institution and Cafeteria Cooks
Waiters
Wholesale and Manufacturing
Sales Representatives
Customer Service
Representatives
Shipping, Receiving, Traffic
Clerks
Laborers, Freight, Stock,
Material Movers
Median Wage
10.27/hour
15